From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from psmtp.com (na3sys010amx143.postini.com [74.125.245.143]) by kanga.kvack.org (Postfix) with SMTP id 8A88B6B0044 for ; Fri, 30 Mar 2012 18:36:42 -0400 (EDT) Received: by iajr24 with SMTP id r24so2095446iaj.14 for ; Fri, 30 Mar 2012 15:36:42 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <20120330153519.1e80735b.akpm@linux-foundation.org> References: <20120330153519.1e80735b.akpm@linux-foundation.org> From: Sasha Levin Date: Sat, 31 Mar 2012 00:36:21 +0200 Message-ID: Subject: Re: mm: migrate_pages hang Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Sender: owner-linux-mm@kvack.org List-ID: To: Andrew Morton Cc: Andrea Arcangeli , Hugh Dickins , Thomas Gleixner , linux-mm@kvack.org, "linux-kernel@vger.kernel.org List" , Dave Jones On Sat, Mar 31, 2012 at 12:35 AM, Andrew Morton wrote: > On Sat, 31 Mar 2012 00:23:30 +0200 > Sasha Levin wrote: > >> Hi all, >> >> I saw the following inside a KVM guest, this is the first time I >> observed this hang and it seems to have started with today's >> (20120330) -next. >> >> [ 3122.093136] INFO: task trinity:17328 blocked for more than 120 second= s. >> [ 3122.093807] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" >> disables this message. >> [ 3122.095237] trinity =A0 =A0 =A0 =A0 D ffff8800703284d8 =A05648 17328 = =A0 3101 0x00000004 >> [ 3122.096543] =A0ffff8800064cfb08 0000000000000082 ffff8800064cfab8 >> ffffffff8107d346 >> [ 3122.098087] =A0ffff8800064cffd8 00000000001d4580 ffff8800064ce010 >> 00000000001d4580 >> [ 3122.099447] =A000000000001d4580 00000000001d4580 ffff8800064cffd8 >> 00000000001d4580 >> [ 3122.100607] Call Trace: >> [ 3122.100983] =A0[] ? kvm_clock_read+0x46/0x80 >> [ 3122.101950] =A0[] ? __lock_page+0x70/0x70 >> [ 3122.102761] =A0[] schedule+0x24/0x70 >> [ 3122.103576] =A0[] io_schedule+0x87/0xd0 >> [ 3122.104371] =A0[] sleep_on_page+0x9/0x10 >> [ 3122.105179] =A0[] __wait_on_bit_lock+0x52/0xb0 >> [ 3122.106042] =A0[] __lock_page+0x62/0x70 >> [ 3122.106811] =A0[] ? autoremove_wake_function+0x40/0= x40 >> [ 3122.107450] =A0[] lock_page+0x39/0x40 >> [ 3122.107976] =A0[] __unmap_and_move+0x274/0x280 >> [ 3122.108848] =A0[] ? list_del+0x11/0x40 >> [ 3122.109385] =A0[] unmap_and_move+0x8d/0x130 >> [ 3122.109957] =A0[] migrate_pages+0xab/0x150 >> [ 3122.110788] =A0[] ? isolate_freepages+0x390/0x390 >> [ 3122.111399] =A0[] compact_zone+0x1e1/0x2a0 >> [ 3122.111935] =A0[] ? _raw_spin_unlock_irqrestore+0x7= 5/0xa0 >> [ 3122.112579] =A0[] __compact_pgdat+0x1b3/0x200 >> [ 3122.113199] =A0[] compact_node+0x37/0x40 >> [ 3122.113631] =A0[] ? lru_add_drain_all+0x10/0x20 >> [ 3122.114266] =A0[] sysfs_compact_node+0x48/0x60 >> [ 3122.114768] =A0[] ? sysfs_write_file+0x82/0xf0 >> [ 3122.115368] =A0[] dev_attr_store+0x1b/0x20 >> [ 3122.115877] =A0[] sysfs_write_file+0x9e/0xf0 >> [ 3122.116423] =A0[] vfs_write+0xc8/0x190 >> [ 3122.116934] =A0[] sys_write+0x4f/0x90 >> [ 3122.117697] =A0[] system_call_fastpath+0x16/0x1b >> [ 3122.118354] 2 locks held by trinity/17328: >> [ 3122.118694] =A0#0: =A0(&buffer->mutex){+.+.+.}, at: >> [] sysfs_write_file+0x44/0xf0 >> [ 3122.119498] =A0#1: =A0(s_active#57){.+.+.+}, at: [] >> sysfs_write_file+0x82/0xf0 > > You reported what I suppose is the same bug a week ago ("mm: hung task > (handle_pte_fault)"): the kernel is waiting for a page to come > unlocked, thinking that there is I/O outstanding against it. > > And my ugh still applies: "There are quite a lot of things which could > cause this, alas. =A0VM, readahead, scheduler, core wait/wakeup code, IO > system, interrupt system (if it happens outside KVM, I guess). =A0So.... > ugh. =A0Hopefully someone will hit this in a situation where it can be > narrowed down or bisected." Oh, I didn't think that this one was related to the one I've previously reported. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: email@kvack.org